xii ARTICULATE IDEAS 303 



In a word, concrete objects are known in their relations, 

 and the knowledge is applied in action. 



These functions being thus closely interconnected, the 

 possession of any one makes the possession of others a 

 matter of some probability. And if there is independent 

 evidence, of some probability, for the possession of each 

 function by the higher mammals, the probabilities stand to 

 strengthen one another. Now, the only evidence to be 

 found for the performance of these or any other functions 

 by animal intelligence, is evidence derived from animal 

 behaviour. Having this in view, we laid down certain points 

 of behaviour by which the practical judgment might be dis- 

 tinguished from the lower stage of intelligence in which 

 action passes out of the control of instinct to come under 

 that of habit. These points were as follows : 



(i). The single Instance. 



Habit except in the direct modification of an impulse 

 is built up by a large number of repetitions. One clear 

 perception of a relation should have a permanent effect. 



On this head, we found abundant evidence that the 

 higher animals frequently learn by a single instance, and 

 normally by what I have called a critical instance, in which 

 their attention seems to be for the first time definitely 

 directed to the facts. 



(2). The Perception of objects in their relations. 



The relations given in concrete experience and guiding 

 subsequent actions may be relations between any objects of 

 perception and are not (as in the case of Assimilation) 

 confined to those between sensation and the motor excite- 

 ments immediately attendant upon them. 



The simplest evidence of the appreciation of such 

 relations is Learning by Perception of Results. A habit 

 can only be acquired by previous action. The practical 

 judgment can be formed on the basis of a perception that 

 the change A leads to the desired result B. 



Of the power of animals to learn in this way, probable 

 though not conclusive evidence has been given. In many 

 cases in which accidental success seems improbable, animals 

 repeated what they had seen done, and in some instances, 

 actions were so co-ordinated as to bring about the physical 



